TW201545819A - Computer vision identification output image assisted led die selection system and selection method - Google Patents

Abstract

The invention relates to a computer vision identification output image-assisted LED die selection system and a selection method. The computer vision recognition output image-assisted LED die selection system comprises: an optical detection device, a scanning device, and a a tablet device and a die picking device, the scanning device scans the detection result of the optical detecting device and transmits the scanned data to the flat panel display device, and then displays the identification signal by the tablet device for the user to quickly identify and simultaneously Through this selection method, it is possible to achieve the effect of reducing labor costs while improving the quality yield and production speed.

Description

Computer vision identification output image assisted led die selection system and selection method

The invention relates to a LED die selection system and a selection method, in particular to a computer vision identification output image assisted LED die selection system and a selection method using a computer tablet output detection result.

As the chip on glass (COG) technology becomes more and more popular, the inspection requirements for COG grains are also growing rapidly. In addition to inspections at the wafer stage, in order to increase production yield and reduce the chance of repair or rework, the inspection of the die placed on the carrier disk before it is mounted to the display panel is also necessary. In order to meet these inspection requirements, a small number of inspection systems for inspecting the dies on the carrier tray were developed.

The use of Automated Optical Inspection (AOI) instruments is also becoming more and more important, using automatic optical inspection for optical image and image comparison detection technology to detect products, and constantly aiming at light, thin, high density electronic Developed for product inspection needs, it can effectively detect various bad processes of components without testing fixtures, electronic measurement, and product destruction.

In addition, the detection of general wafers is mainly divided into two stages. The first stage is the inspection of the whole wafer. The wafer is set on the automatic optical inspection instrument to detect in a wide range of scanning, and then manually. The method is further performed by scanning electron microscope re-detection, and the 瑕疵 wafer sample detected by the automatic optical detecting instrument is further judged, and in the case of detecting the crystal grain, it is after cutting The crystal grains are manually detected by an electron microscope to pick out defective crystal grains, but the cost of detecting the manual inspection is greatly increased, firstly, a lot of labor costs are required, and manual processing of the crystal grains is performed. It takes a lot of education and training to effectively use the visual method to know the defective crystal grains. Moreover, there are many objects using the crystal grains, and the corresponding demand conditions are relatively complicated, and the overall quality yield of the manual detection method. The conventional technology has the following disadvantages: 1. The cost of the detection is high; 2. The overall quality yield cannot be effectively maintained; 3. The detection operation speed is low.

Therefore, how to solve the above problems and problems in the past, that is, the inventors of this case and the relevant manufacturers engaged in this industry are eager to study the direction of improvement.

Accordingly, in order to effectively solve the above problems, the main object of the present invention is to provide a computer vision recognition output image assisted LED die selection system which can effectively reduce labor costs and at the same time improve quality yield and production speed.

A secondary object of the present invention is to provide a computer vision identification output image assisted LED die selection system that can effectively reduce labor costs while improving quality yield and production speed.

To achieve the above objective, the present invention provides a computer vision identification output image assisted LED die selection system, comprising: an optical detection device, a scanning device, a tablet device, and a die suction device, wherein the optical The detecting device mainly has at least one carrying fixture, and the carrying fixture can be provided with a plurality of wafer positioning modules, wherein the wafer positioning modules are respectively positioned with a wafer, and each of the wafers has a plurality of LED dies. And the optical detecting device detects the crystal via Rounding and generating at least one wafer report, the wafer report includes a code and a detection data; the scanning device scans a barcode of the wafer report, and generates a scan data via the barcode, and the tablet device is electrically Connected to the scanning device and receive the scanning signal, and the tablet device has a display screen, and the detected wafer is disposed on the display screen, and the display screen generates at least one according to the scanned data. Displaying a signal and an identification signal, wherein the display signal is aligned with the wafer, and the identification signal is displayed at the position of the LED die that has not passed the detection, and the die picking device picks up the position where the identification signal is generated. An LED die, whereby the computer vision recognition output image-assisted led die selection system scans the detection result of the optical detection device by using the scanning device, and transmits the scan data to the electrically connected flat panel display device, And the display screen of the tablet device generates an identification signal at the position of the failed LED die for the user to use the identification signal. Directly judging the LED dies that have not passed the detection, and then directly drawing the undetected LED dies by the dies suction device, which can effectively solve the steps required to be selected by the human through the microscope, thereby reducing the labor cost and simultaneously It can improve the quality of the product and the speed of production.

The present invention further provides a computer vision identification output image-assisted LED die selection method, the method comprising: providing a plurality of wafers, the wafer having a plurality of LED dies; and locating a plurality of wafers of the plurality of wafers The positioning module is disposed on a bearing fixture; the bearing fixture is disposed in an optical detecting device for detecting and removing and removing the wafer after the detecting; the optical detecting device generates at least one wafer report. Scanning the bar code of the wafer report by a scanning device and generating a scan data; providing a tablet device to receive the scan data and generating a display signal according to the scan data and generating a position of the LED die that fails the detection An identification signal, and displaying the display signal and the identification signal on a display screen; the wafer is disposed on the display screen and Pairing the display signal; picking up the LED die at the position where the identification signal is generated by a die picking device.

Thereby, the computer vision identification output image-assisted LED die selection method can scan the detection result of the optical detection device by using the scanning device, and transmit the scan data to the flat panel display device, and display the display screen of the tablet device An identification signal is generated at the position of the LED die that has not passed the detection, and when the wafer is disposed on the display screen, the user can directly determine the position of the LED die that has not passed the detection by the identification signal, and then The die-absorbing device directly absorbs the undetected LED die, which can effectively solve the steps required to be selected by the human through the microscope, thereby achieving the effect of reducing the labor cost and simultaneously improving the quality yield and the production speed.

The above object of the present invention, as well as its structural and functional features, will be described in accordance with the preferred embodiments of the drawings.

Please refer to FIGS. 1 and 2 and FIGS. 3 and 4, which are block diagrams and implementation diagrams 1 and 2 and FIG. 3 of the first preferred embodiment of the present invention, wherein the computer vision The identification output image assisting LED die selection system comprises: an optical detecting device 2, a scanning device 4, a tablet device 5 and a die picking device 6; wherein the optical detecting device 2 has at least one carrying device The carrier fixture 21 can be provided with a plurality of wafer positioning modules 22, and each of the wafer positioning modules 22 has a fixed wafer 3, and each wafer 3 has a plurality of LED dies. 31. After the carrier 3 is mounted on the carrier 3, the carrier fixture 21 is sent to the optical detecting device 2, and the optical detecting device 2 can be used. The wafer 3 of each wafer positioning module 22 is detected to detect the LED die 31 having germanium, and the detected wafer 3 is removed by the optical detecting device 2 and is removed by the carrying fixture 21, and then The fixed wafer 3 is separated from the wafer positioning module 22.

Please refer to FIG. 1 and FIG. 5 and FIG. 6 , which are block diagrams and implementation diagrams 4 and FIG. 5 of the first preferred embodiment of the present invention, wherein the optical detecting device 2 is directed to the crystal After the round 3 detection, a wafer report 23 is generated, each wafer report 23 is generated for each wafer 3, and the wafer report 23 includes one barcode 231 and one inspection data of each wafer 3. 232, and the scanning device 4 scans the barcode 231 of the wafer report 23, and generates a scan data 41 via the barcode 231, and the scan data 41 is transmitted to the tablet device 5, and the tablet computer The device 5 has a processing unit (not shown), and the tablet device 5 receives the scanned data 41 and then converts through the processing unit, and displays a display on one of the display devices 51 of the tablet device 5. The signal 511 and an identification signal 512, wherein the display signal 511 is generated in a range equal to the range of the wafer 3, and the detached wafer 3 is aligned, and the identification signal 512 is For the display screen 51 is generated Different light sources.

Please refer to FIG. 1 and FIG. 7 and FIG. 8 , which are block diagrams and implementation diagrams 6 and FIG. 7 of the first preferred embodiment of the present invention, wherein the display screen 51 of the tablet device 5 is shown. After the display signal 511 and the identification signal 512 are generated, the removed wafer 3 can be disposed on the display screen 51, and the wafer 3 can be aligned via the display signal 511. After the circle 3 is positioned, some of the LED dies 31 of the wafer 3 generate a light source via the identification signal 512, and the LED dies 31 that generate the light source are LED dies having defects detected by the optical detecting device 2. 31, that is, the identification signal 512 first refers to the relative position of the display signal 511 and the wafer 3, and generates a light source for the position of the undetected LED die 31, and is positioned at the wafer 3. After that, the LED die 31 that generates the light source is also the LED die 31 that is not detected by the optical detecting device 2, and the display screen 51 does not generate the identification signal 512 at the position of the detected LED die 31. The position of the LED die 31 that has not passed the detection can be directly determined by the identification signal 512, and the LED die 31 that has not passed the detection is directly sucked by the die picking device 6, thereby effectively solving the requirement of the wafer 3. It is possible for humans to select whether or not there is a step through the microscope, thereby effectively achieving the effect of reducing labor costs while improving the quality yield and production speed.

Please refer to FIG. 9 , which is a schematic diagram of the implementation of the second preferred embodiment of the present invention. The preferred embodiment is substantially the same as the first preferred embodiment and the same function, wherein the main difference is the embodiment. The system 1 includes a control device 7 electrically connected to the tablet device 5 for its wafer 3 (see page 7). When the display screen 51 is disposed on the display screen 51 and is aligned via the display signal 511, the position of the display signal 511 and the identification signal 512 on the display screen 51 can be controlled by the control device 7.

FIG. 10 is a schematic view showing the implementation of the third preferred embodiment of the present invention. The preferred embodiment is substantially the same as the first preferred embodiment. The main difference of this embodiment is that the wafer report 23 generated by the optical detecting device 2 is uploaded to a network server 8, and the tablet device 5 downloads the wafer through the network to its network server 8. Statement 23.

Referring to FIG. 11 and FIG. 1 together, it is a flowchart of a first preferred embodiment of the selection method of the present invention. The computer visual identification output image assisting LED die selection method comprises the following steps:

S1: providing a plurality of wafers, the wafer having a plurality of LED dies; A plurality of wafers 3 are provided therein, and each of the wafers 3 has a plurality of LED dies 31.

S2: The plurality of wafer positioning modules positioned with the plurality of wafers are disposed on a bearing fixture; and the wafer positioning module 22 positioned with the wafers 3 is disposed on the bearing fixture 21.

S3: the carrier fixture is set in an optical detecting device for detection and removed after the detection and the wafer is removed; the bearing fixture 21 is sent to the optical detecting device 2, and the optical detecting device 2 can be The wafer 3 of each wafer positioning module 22 is detected to detect the LED die 31 having germanium, and the detected wafer 3 is removed by the optical detecting device 2 and is removed by the carrying fixture 21, and then Separating the fixed wafer 3 from the wafer positioning module 22

S4: The optical detecting device generates at least one wafer report, scans the barcode of the wafer report by a scanning device and generates a scan data; and the optical detecting device 2 generates a crystal after detecting the wafer 3. The round report 23, each of which is generated for each wafer 3, and the wafer report 23 includes the barcode 231 of the wafer 3 and the associated detection data 232, and the scanning device 4 Scanning the barcode 231 of the wafer report 23, and generating the scanned data 41 through the barcode 231.

S5: providing a tablet device to receive the scan data, generating a display signal according to the scan data, generating an identification signal for the undetected LED die position, and displaying the display signal and the identification signal on a display On the screen, the scanned data 41 is transmitted to the tablet device 5, and the tablet device 5 receives the scanned data 41 and then converts through the processing unit and displays on the display screen 51 of the tablet device 5. There is a display signal 511 and an identification signal 512, wherein the display signal 511 is alignable for the unloaded wafer 3, and the identification signal 512 is different on the display screen 51. light source.

S6: the wafer is disposed on the display screen and disposed on the display signal; the wafer 3 on which the wafer is removed is disposed on the display screen 51, and the wafer 3 can be processed via the display signal 511. Counterpoint.

S7: picking up the LED die at the position where the identification signal is generated by a die picking device.

The LED dies 31 of the wafer 3 generate a light source via the identification signal 512, and the LED dies 31 that generate the light source are the LED dies 31 detected by the optical detecting device 2, and then the dies. The pick-up device 6 directly picks up the LED dies 31 where the identification signal 512 is generated.

Therefore, the user can directly determine the position of the LED die 31 that has not passed the detection by the identification signal 512, and then directly suck the undetected LED die 31 by the die picking device 6, thereby effectively solving the wafer 3. It is necessary for humans to select whether or not there are steps through the microscope, thereby effectively achieving the effect of reducing labor costs while improving the quality yield and production speed.

Referring to FIG. 12 and FIG. 1 together, it is a flowchart of a second preferred embodiment of the selection method of the present invention. The preferred embodiment is substantially the same as the first preferred method of the foregoing selection method and its function. The same, wherein the main difference of the embodiment is the step S4, the step S4: the optical detecting device generates at least one wafer report, and the detector uploads the wafer report to a network server, and the tablet The computer device downloads the wafer report through the network to its network server, scans the barcode of the wafer report with a scanning device, and generates a scan data; wherein the optical detecting device 2 generates the wafer The report 23 is uploaded to a web server 8, and the tablet device 5 downloads the wafer report 23 via the network to its web server 8.

Further, the tablet device 5 is installed with an application (Application, APP), and The wafer report 23 is downloaded via the network to its web server 8 via the application.

Referring to FIG. 13 and FIG. 1 together, it is a flowchart of a third preferred embodiment of the selection method of the present invention. The preferred embodiment is substantially the same as the first preferred method of the foregoing selection method and its efficacy. The main difference is that the display signal 511 and the identification signal 512 are displayed on the display screen 51. The step S6: controlling the display signal 511 and the identification signal 512 to be displayed through a control device 7. The screen 51 is displayed at a position; and the step S7 is: the wafer is disposed on the display screen and is located on the display signal; and step S8: the position of the identification signal is generated by a pair of die picking devices. The LED die is removed.

As described above, the present invention is a computer vision identification output image assisted LED die selection system And the selection method has the following advantages: 1. reducing the cost; 2. improving the quality yield; 3. increasing the production speed.

It is to be understood that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and changes may be made without departing from the spirit of the invention, for example, for configuration or arrangement. The type is transformed, and the equivalent effect of various changes, modifications and applications shall be included in the scope of the case and shall be combined with Chen Ming.

In summary, the computer vision identification output image assisted led die selection system and the selection method of the present invention can achieve the effect and purpose when used, so the invention is a practical and excellent creation, in accordance with the invention. The application requirements of the patent, 提出 apply in accordance with the law, and hope that the trial committee will grant the case as soon as possible to protect the inventor's hard work. If there is any doubt in the arbitral tribunal, please do not hesitate to give instructions. The inventor will try his best to cooperate with him.

1‧‧‧Computer vision identification output image assisted led die selection system

2‧‧‧Optical inspection device

21‧‧‧bearing fixture

22‧‧‧ Wafer Positioning Module

3‧‧‧ wafer

31‧‧‧LED dies

23‧‧‧ Wafer Report

231‧‧‧ barcode

232‧‧‧Test data

4‧‧‧Scanning device

41‧‧‧ scan data

5‧‧‧ tablet device

51‧‧‧ Display screen

511‧‧‧Display signal

512‧‧‧ identification signal

6‧‧‧Grad suction device

7‧‧‧Control device

8‧‧‧Web server

1‧‧‧Computer vision identification output image assisted led die selection system

2‧‧‧Optical inspection device

23‧‧‧ Wafer Report

231‧‧‧ barcode

232‧‧‧Test data

3‧‧‧ wafer

31‧‧‧LED dies

4‧‧‧Scanning device

41‧‧‧ scan data

5‧‧‧ tablet device

51‧‧‧ Display screen

511‧‧‧Display signal

512‧‧‧ identification signal

6‧‧‧Grad suction device

Claims (10)

A computer vision identification output image-assisted LED die selection system includes: an optical detection device having at least one bearing fixture, wherein the carrier fixture is provided with a plurality of wafer positioning modules for positioning a wafer, the wafer Having a plurality of LED dies, and the optical detecting device generates at least one wafer report by detecting the wafer, the wafer report includes a code and a detection data; and a scanning device that reads the crystal according to the barcode Circular report and generating a scan data; a tablet device having a display screen and a wafer set for detection, and the tablet device is electrically connected to the scan device and receiving the scan data, and the display screen Generating at least one display signal according to the scan data and generating an identification signal for the undetected LED die position; and a die picking device for picking up the LED die that generates the identification signal. The computer vision identification output image-assisted LED die selection system according to the first aspect of the invention, wherein the tablet device has a processing unit, and the processing unit generates the display signal and the identification signal. The computer vision identification output image assisted LED die selection system according to claim 1, wherein the die suction device is a vacuum suction device. The computer vision identification output image-assisted LED die selection system of claim 2, wherein the identification signal is not generated at the LED die position of the display screen through the detection. The computer vision identification output image-assisted LED die selection system according to claim 1, further comprising a control device electrically connected to the tablet computer The display signal and the identification signal are controlled and displayed on the display screen. The computer vision identification output image-assisted LED die selection system according to claim 1, wherein the optical detection device generates the wafer report and uploads to a network server, and the tablet device Download the wafer report via the network to its web server. A computer vision identification output image-assisted LED die selection method includes the steps of: providing a plurality of wafers having a plurality of LED dies; and locating a plurality of wafer positioning modules of the plurality of wafers The carrier fixture is disposed in an optical detecting device for detecting and removing the wafer after the detecting and removing the wafer; the optical detecting device generates at least one wafer report to the scanning device Scanning the bar code of the wafer report and generating a scan data; providing a tablet device to receive the scan data and generating a display signal according to the scan data and generating an identification signal for the undetected LED die position, And displaying the display signal and the identification signal on a display screen; the wafer is disposed on the display screen and located opposite to the display signal; and the position of the identification signal is generated by a pair of die picking devices The LED dies are picked. The computer vision identification output image-assisted LED die selection method according to claim 7, wherein the step of generating the wafer report by the optical detecting device further comprises: the detector uploading the wafer report thereof Up to a web server, and the tablet device downloads the wafer report through the network to its web server. Computer vision identification output image assisted led die as described in claim 8 A method of selecting, wherein the tablet device is installed with an application and downloads its wafer report via the network to its web server via the application. The computer vision identification output image-assisted LED die selection method according to the seventh aspect of the invention, wherein the display signal and the identification signal are displayed after the display screen has a step: controlling the The display signal and the identification signal are displayed on the display screen.